|Publication number||US7180424 B2|
|Application number||US 10/528,072|
|Publication date||Feb 20, 2007|
|Filing date||Nov 5, 2003|
|Priority date||Nov 8, 2002|
|Also published as||EP1570448A2, EP1570448A4, US20060170561, WO2004042670A2, WO2004042670A3|
|Publication number||10528072, 528072, PCT/2003/918, PCT/IL/2003/000918, PCT/IL/2003/00918, PCT/IL/3/000918, PCT/IL/3/00918, PCT/IL2003/000918, PCT/IL2003/00918, PCT/IL2003000918, PCT/IL200300918, PCT/IL3/000918, PCT/IL3/00918, PCT/IL3000918, PCT/IL300918, US 7180424 B2, US 7180424B2, US-B2-7180424, US7180424 B2, US7180424B2|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (6), Referenced by (9), Classifications (18), Legal Events (6)|
|External Links: USPTO, USPTO Assignment, Espacenet|
This Application is the National Stage of International Application No. PCT/IL 03/00918 filed Nov. 6, 2003, which claims the benefit of U.S. Provisional Application No. 60/424,725 filed Nov. 8, 2002. Additionally, this application claims the benefit of Israeli Patent Application No. IL164086, filed Sep. 14, 2004.
The present invention relates to the field of premises security systems, and more specifically, to a computerized security system which incorporates the ability to interpret canines' (dogs) behavior. The use of dogs in securing premises is well known in the art since dogs have the natural ability to detect and react to suspicious activities in their vicinity.
There are many advantages to using dogs for securing premises besides having excellent detection abilities; dogs constitute a physical and a psychological deterring factor merely by their presence, or even just by the knowledge of their existence. In addition to providing a visible presence that deters potential criminal activity, they can abort such activities and detain any persons involved. Dogs may also be trained for specific and even highly complex security missions and can therefore answer special security needs.
In spite of their many advantages, dog-based security systems also have their shortcomings. For a dog-based security system to be effective, security personnel need to be in reach of the dogs for constant check-up and within hearing distance. For this end some systems make use of close-circuit surveillance systems providing full visual and audio coverage of the secured areas, which makes them costly and complex systems.
An additional problem with dog-based security systems is that dogs tend to bark for different reasons, not always for the purpose of indicating threat. Dogs may bark due to the appearance of an animal, at other dogs or at a friendly caretaker. Frequent barking may, at first, cause many false alarms and, in time, may cause the guards to unintentionally decrease their level of alertness.
A third problem stems from the fact that the dogs are vulnerable to direct attacks by weapons or by use of toxins—they can be shot and killed, drugged or poisoned. For such scenarios is required a Dog-based security systems which rely upon direct and constant visual contact between the guards and the dogs.
There is therefore a need for security systems that integrate automatic means into dog-based security methods. Such systems can then be attentive to indications given by the dogs and analyze them into readings of level of alarm. These systems then also monitor the state of the dogs on watch, and can activate an alarm whenever an attempt to hurt the dogs is made.
New developed applications for interpreting dogs' behavior are designed for the purpose of improving the communication between pet-dogs and their owners. Such a device is disclosed in Japanese Patent No. JP3083915, which attempts to interpret a dog's barks and cries and translate it to predetermined expressions of emotions and needs. The device includes a voice pattern for each dog type. Based on comparing dog barking patterns to a database of barks, the device is supposed to “translate” the dog's vocal expressions into a range of terms.
This patent is particularly aimed for leisure purposes not requiring high reliability of identification and is thus unsuitable for security systems, which require accurate recognition of the dog's sensed level of alarm. Furthermore, there are some major and fundamental technological differences:
There is therefore a need for a computerized dog-based security system that can translate vocal and physiological indications given by dogs, to the security alarm level.
A security system and method for alerting of security situations based on watchdogs' behavior which operates according to collected and analyzed auditory and physiological data from the dogs. This system is comprised of sensors for sampling dog sounds in the watchdog's vicinity, of means for collecting physiological data attached to the dog's body or implanted in it, and of means for processing the collected data and for identifying an alert situation. The audio data is converted to digital form and amplified by analog to digital converter and amplifier. The system then differentiates between different signal sources and identifies single barks sounds in accordance with received measurements. The system is also comprised of a means for identifying characteristic sound signal patterns and determining alert level by comparing the sound patterns to predefined values. Physiological data which include heartbeat rate, rate of breathing, muscle tonus, blood pressure and skin conductivity is also compared to predefined data and the dogs' state is estimated. According to valuated data the system assesses the level of security sensed by the dogs and in case of an alarming situation the user is notified via a security terminal unit.
In order to analyze the pattern of barking the system first filters background noises, distinguishes between barks from different sources and normalizes the signals. The sampled barks are then analyzed according to the characteristics of each single bark, such as the bark's frequency and pitch, and according to their characteristics over time such as measuring the mean and variant of time intervals, bark rate variability, barking rate and the distribution variance of the barks. The analyzing means may be embedded in the sensor means in the dogs' vicinity or remote, and the system therein includes of transmitting facilities for enabling connectivity between its different components.
The present invention is a computerized dog-based security system which receives auditory and/or physiological signals from a single or a group of dogs on guard, analyzes them and gives indications as to the alert level according to the dogs' emotional state at any given point in time. The invention is a new and efficient means for integrating the many benefits and advantages of dogs (dogs) into an automatic interpretive security system.
The system receives auditory and/or physiological signals from dogs, performs a unique digital signal processing analysis, and provides an indication as to one of several possible system alert levels and events. At the heart of the system is a series of proprietary audio and physiological data processing algorithms, which digitally sample the signals, filter, identify and analyze the data profile according to predetermined parameters. It can detect the vitality signs of any specific dog, determine whether it is alive and awake, and measure its level of alertness. Based on this analysis, the system can provide a highly accurate estimation of the dogs' state of alertness in a specific location.
The preferred embodiment of the present invention as illustrated in
The processing and analyzing unit 200 is comprised of a signal receiver 210, an analogue to digital (A/D) converter and amplifier 220, a central processing unit 230 and a communications interface 240. The signal receiver 210 receives the signals from each sensor unit 100 and transfers these signals to the A/D converter 220. The converter 220 amplifies the analog signals and transforms the analog signals to digital signals. A central processing unit 230 executes the main processing and analyzing algorithms. It includes an audio processing and analyzing algorithm 231, a physiological processing and analyzing algorithm 232 and a security decision module 233.
The audio processing and analyzing algorithm 231 receives the auditory signals and analyses them, hence providing a calculated estimation of the dog's psychological condition (indicating the dogs' alertness level). As illustrated in the flow charts in
The physiological analyzing algorithm 232 processes (see
The security decision algorithm module 233 processes (see
Thereafter, the system determines if the current time frame is a qualified frame by checking whether in the current time frame the signal contains any valid sound patterns. The decision is based on the output of the BPFs' energy levels. This saves analyzing resources since it prevents the system from redundant operation if the signal contains only silence or background noise. If the frame is announced as having substantial data the process continues to the next stage.
Block 231 b in
The next step estimates the pitch or the fundamental frequency of a bark. In addition, it looks for stress signs in the dog's bark.
Three major features of the sampled signal are analyzed on the basis of pitch measurement: distinguishing between barks originating in response to threat and barks, which are a response to routine activity (such as play, trainer, patrol or a stray cat, etc.); distinguishing between barks coming from different dogs and ruling out any sounds which are not barks, such as human speech or shouts. If a certain bark successfully meets all of the requirements of a valid bark, it is qualified and it remains in the bark vector. Otherwise, it is eliminated from the vector.
Next, spectral and cepstral analysis are performed for each qualified bark, and each bark is classified as one of three possible alert levels. The classification is performed by comparing the bark's spectral and cepstral features to a set of statistical models.
Then, clustering may be performed. The purpose of clustering is to classify the barks in the bark vector into groups that correspond to the possible sources of the audio signal. This classification is mainly done based on their pitch but it may also include other features. The source separation enables to distinguish between barks from different dogs. The clustering process further contributes to eliminating noise such as music, and human voices.
Finally, the clustered bark vector is analyzed statistically, calculating: the intervals between single barks, mean and variance of barking rate and barks distribution.
This statistical data, combined with the pitch, spectral, cepstral and stress features are delivered to the final decision module 233. The statistical decision module 233 receives features and parameters that characterize emotional and physiological behavior of the current time frame. The decision may result in determining that there is no alert, that the alert is at a medium level or that there is a high alert. Alert at a medium level might mean that there is some suspicious activity that needs attention, and high alert means that there is a definite territory intrusion, which may be an indication of a security breach.
At the same time the system can analyze every dog's physiological indicators, as received from the physiological sensor unit 120 as illustrated in
Integrating the information given by the analysis of the dog's auditory signals and its physiological status gives a highly accurate representation of the security status as it is sensed by the dog.
Monitoring the physiological status of the dog also ensures having a constant update on the well-being of the dog and his performance relating to security maters. Any attempt to hurt a dog, whether successful or unsuccessful, will be immediately apparent and would activate the alarms.
The user security terminals 300 which are illustrated in
Additionally, the system can be easily adjusted to address specific needs, by training the dogs to react in a certain manner to specific situations and by translating the respective auditory and physiological indications of these reactions accordingly. By way of example, if, the dogs are trained to react intensively to an approaching vehicle as opposed to any person approaching by foot, they might still bark if a person appears, but their reaction will be a moderate one and it will be interpreted as such by the system, whereas an approaching car will cause the dogs to give a stressful reaction and activate the alarm.
While the above description is very specific, these should not be construed as limitations on the scope of the invention, but rather as exemplifications of the preferred embodiments. Those skilled in the art will envision other possible variations that are within its scope. Accordingly, the scope of the invention should be determined not by the embodiment illustrated, but by the appended claims and their legal equivalents.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US5818354 *||Apr 28, 1997||Oct 6, 1998||Gentry; Timothy G.||Animal monitoring system|
|US6058889 *||Apr 15, 1998||May 9, 2000||Innotek Pet Products, Inc.||Combination confinement system and bark inhibitor|
|US6535131 *||Aug 23, 1999||Mar 18, 2003||Avshalom Bar-Shalom||Device and method for automatic identification of sound patterns made by animals|
|US6761131 *||Jun 20, 2003||Jul 13, 2004||Index Corporation||Apparatus for determining dog's emotions by vocal analysis of barking sounds and method for the same|
|US20020026311 *||Oct 18, 2001||Feb 28, 2002||Yamaha Corporation||Apparatus for responding to sound at a remote terminal|
|US20030130568 *||May 31, 2002||Jul 10, 2003||Vodyanoy Vitaly J.||Apparatus and method for the measurement of the aerodynamics of olfaction in animals and man|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US7800506||Sep 21, 2010||Eyal Zehavi||Canine security system|
|US8677941 *||Sep 13, 2012||Mar 25, 2014||Faunus Ltd.||System and methods for health monitoring of anonymous animals in livestock groups|
|US9185885||Oct 1, 2014||Nov 17, 2015||Forget You Not, LLC||Digital activity center for pets|
|US20050068183 *||Sep 21, 2004||Mar 31, 2005||Tomokazu Yuasa||Security system and security method|
|US20050197546 *||Mar 3, 2005||Sep 8, 2005||Eitan Mardiks||Security system|
|US20080018481 *||Dec 27, 2006||Jan 24, 2008||Eyal Zehavi||Canine security system|
|US20100198024 *||Aug 5, 2010||Ron Elazari-Volcani||Vitality meter for health monitoring of anonymous animals in livestock groups|
|US20130006065 *||Sep 13, 2012||Jan 3, 2013||Faunus Ltd.||System and methods for health monitoring of anonymous animals in livestock groups|
|US20150022329 *||Jul 16, 2013||Jan 22, 2015||Forget You Not, LLC||Assisted Animal Communication|
|U.S. Classification||340/573.1, 340/870.16, 600/300, 340/539.11, 119/51.02, 340/573.2, 600/303, 119/174|
|International Classification||G08B13/16, G08B23/00, G08B1/08, G08B13/22|
|Cooperative Classification||G08B13/1672, G08B1/08, G08B13/22|
|European Classification||G08B1/08, G08B13/16B2, G08B13/22|
|Mar 17, 2005||AS||Assignment|
Owner name: BIO SENSE TECHNOLOGIES (B.S.T) LTD., ISRAEL
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:EYAL, ZEHAVI;REEL/FRAME:017052/0103
Effective date: 20050227
|Sep 27, 2010||REMI||Maintenance fee reminder mailed|
|Nov 24, 2010||FPAY||Fee payment|
Year of fee payment: 4
|Nov 24, 2010||SULP||Surcharge for late payment|
|Sep 29, 2014||FPAY||Fee payment|
Year of fee payment: 8
|Sep 29, 2014||SULP||Surcharge for late payment|